#
W-0165a
INTERNATIONAL
MOTOR VEHICLE PROGRAM
FY ’97 IMVP WORKING PAPERS
CREATING
LEAN SUPPLIERS:
PRODUCTION
DIFFUSING
LEAN
THROUGH THE SUPPLY CHAIN
John Paul MacDuffie and Susan Helper
Creating Lean Suppliers: Diffhsing Lean Production Through the Supply Chain
John Paul MacDufile Wharton Schoo~ University of Pennsylvania Susan Helper Case Western Reserve University
July 1997
Final Draft Not for Quotation or Distribution Without Permission
John Paul MacDufHc 2017 Steinberg-Dktrich Hall Wharton School University of Pennsylvania Philadelphia PA 19104-6370 (215) 898-2588 (tel) (215) 898-0401 (f=)
[email protected]. edu
Susan Helper Center for Regional Economic Issues Case Western Reserve Univemity 10900 Euclid Avenue Clevelani OH 44106 (216) 368-5541 (tel) (216) 368-5542 (f=)
[email protected]
We are gratefid to Honda of American Manufacturing and the Honda suppliers who generously provided time and access to their operations for this project. Thanks also to Paul Adler, Robert Cole, Mark FndQ Martin Kenney, Ann Marie Knot& David Levine, Charles Sabei and participants in Wharton’s Organizational Learning Seminar for comments on an earlier draft. Funding for this research was provided by the International Motor Vehicle Progam at M.I.T., the Jones Center for Management Policy, Strategy, and Organization at Whartou and the Center for Regional Economic Issues at Case Western Reserve University. :.$ .,,,,: ;+
Page 1
Hon& of America has developed a comprehensive approach to teaching the principles of lean production to its suppliers. The centerpiece of these efforts is a program called BP (for “Best process”, “Best Performance “, “Bs Practice”), in which a crossfunctional team of persomel born Honda and the supplier work intensively for week or even months on narrowly-targeted improvement projects in the supplier’s plant. BP has been quite successfid in enhancing supplier performance; suppliers participating in the program in 1994 avmge~ productiviw gains of 50°/0 on lines reengineered by BP. However, Honda found there was high variation in the extent to which suppliers were able to transfer the lessons taught beyond the line or plant where the BP intervention occurred. We explore the reasons for this variatio~ touching on how the BP process interacts with the broader relationship between customer and supplier, organizational learning, technology transfer, and the transplantation of Japanese management practices to the U.S. The case studies we present of three of Honda’s U.S. suppliers illustrate the dynamics of the learning process and the complex relationship that emerged between “teacher” and “student”. We found that achieving self sufficiency with the lean production techniques taught by BP is more likely when the supplier has a moderate degree of identification with and dependency on the customer. If these are too hi~ the supplier will be tempted to continue to rely on the customer for assistance; if they are too low, the learning relationship may break down. It appears that Honda has achieved the most supplier self reliance with larger U.S.-owned companies, who have an identity as strong, competent actors, and thus try to reduce dependence on Honda by mastering the new knowledge quickly. Yet these larger suppliers may be less responsive to Honda’s needs that small-to-medium suppliers whose capabilities can be boosted through Honda’s supplier development activities.
Page2 The existence of supplier+ustomer relationships, piuticulady among Japanese companies, that fit neither “market” nor “hierarchy” categories has drawn sustained attention in recent years (Helper, 1991, 1992; Smit@ 1992; Nishiguc~
1994; Dyer and Ouchi, 1993; Cusurnano and Takeishi, 1995). Such
amngements are integral to the structm
of Japanese business networks (Gerlaclq 1993) and they have
also been a companion to Japanese direct investment in overseas manufacturing fmilities.
In the auto
industry, Japanese companies initially used parts from Japan in the new assembly plants built in North America in the 1980s. But since thenj they have steadily increased the percentage of parts purchased from U.S.-located suppliers of both U.S. and Japanese ownership (Mair, 1993; Kemey and Florid% 1993). III the process, some of these Japanese companies have taken the unusual step of working extensively with their suppliers to teach them “lean production” - ofien by sending their own employees into supplier plants for weeks or months to redesign work stations, reorganize process flow, modifi equipment, and establish problem-solving groups. This level of involvement with the internal operations of externally-owned firms is unprecedented and raises the qmtion
“why create lean suppliers?” One answer is that Japanese companies have not
been able to continue supplying their U.S. assembly plants from Japan-based suppliers, because of political pressure to source parts locally (due to Japan’s persistent trade surplus with the U.S.) and the strong economic incentive to move production overseas provided by the strength of the yen. Yet as we discuss below, this is not sufficient to explain the intensity of the supplier support activities undertaken most extensively by Hon& Toyo@ and Nissan (Bennet, 1994; Florida and Jenkins, 1996). in this paper, we examine Honda’s supplier support effo~ in particular, drawing upon extensive field work at six Honda suppliers. 1 Case studies of thee of these suppliers are featured here; Table 1 identifies their key characteristics. We spent a total of eight days at HonA and from one to two days at each of the supplier sites. 2 We asked respondents (who included purchasing and supplier support staff at Honda as well as managers, shopfloor workers, supemisom, union officials, production engineers, and corporate staff at the suppliers) to tell us about key events in the business relationship with Honda but also about “problems, issues, and opportunities for improvement” that emerged over the course of their relationship with Honda We begin by considering the question raised earlier - “why create lean supplied?”. review Honda’s philosophy of supplier relations and intioduce the “BP” pro-
Next we
perhaps Honda’s central
supplier support activity. “BP” has many meanings - best practices, best process, best profits - and a
Page3 distinctive approach to transf-g
technical knowledge related to lean production. At the heart of the
paper are the three BP case studies that reveal the complex dynamics accompanying this mechanism for knowledge transfer. We then analyze the impact of BP on supplier capabilities along various dimensions. Finally, we utilize concepts from research on organizational learning to draw out the general implications of Honda’s BP - for customers who want to boost the performance of their supply chain by providing technical assistanq
for suppliers who must try to absorb this knowledge while coping with customers
who want access to their internal operations; and for understanding what sorts of “mediating mechanisms” for knowledge transfer are most likely to yield suppliers who are self-sufficient and capable rather than dependent on their customer for ongoing support.
Why Create Lean Suppliers? If one thing is clear tim half a century of management research (and the experience of countless companies), it is that organizational change is difficult to bring about and even more difficult to sustain (Dosi and Ko~
1992; Nelson and Winter, 1978; Hannan and Freem~
1984). Given this, it is by no
means obvious that a company should undertake to bring about organizational change at its suppliers. Lean production in the auto industry, as described by M.I.T.’s International Motor Vehicle Program (with which we are both affiliated), involves fu-reaching organizational and technological changes. Within a fhn’s own manufiwturing opemtiou it involves reducing bu.t%rs through Just-in-Time inventory systems, producing only what is needed by downstream “customers”, whether internal or external; pushing down responsibilities for quality inspection and the specification of work tasks to motivat@ muitislcilled workers organized into teams; eliciting a steady stream of ideas for process improvement (kuizen) from employees at all levels (Womac~ Jones, and Roos, 1990). Added to this, customers are likely to demand that suppliers assume substantial responsibility developm-,
during product
accommodate customer rquests for engineering changes in their product or manufacturing
process; be highly reliable with respect to qualityand delivery; and have the ability to respond quickly in case of problems. These requirements are difficult for a supplier to meet unless they have adopted lean production practices themselves. Thus a lean customer is likely to find it more productive to work with lean suppliers. However, the adoption process can be rislq, since it is common for improvements on one dimension (e.g. reducing inventory levels) to have the initial impact of reducing performance on another
Page4 dimension (e.g. delivery reliability or responsiveness to customer schedule changes).
Thus where an
automaker possesses an important capability such as lean production that it wants to establish in its supply base, several alternatives might appear more appealing than trying to develop those capabilities among not-lean suppliers. Do it yoursel~
Vertical integration was once the clear preference when manufiwturing firms
wanted to insure that they controlled the output of upstream processes. Presumably the firm understands its input requirements better than anyone else, already has manufacturing capabilities, can obtain economies of scale not available to a more decentralized supply base, and can insure through . . dmmtmtive fiat that input prices will not be monopolistic. More recently, vertical integration has fh.llen out of favor, due in part to the advantages of long-term relationships with separate supplier companies demonstrated by Japanese companies (Nishiguchi, 1994; Smitlq
1991).
If parts are single or
dual-source~ suppliers maybe able to achieve substantial economies of scale. The customer can help the supplier with technical assistance while not bearing full investment costs, and can still benefit from any supplier improvements due to the stipulation that productivity gains will be shared. By focusing on a single product line, suppliers can develop innovations that are beyond the customer’s ability to achieve. Extensive tacit knowledge can develop in the supplier-customer relationship, facilitating coordination of the respective expertise of the parties, particularly with respect to complex value-added tasks such as product development. Switch to a lean supplier. If a lean customer can arrange to do business with suppliers who are already leq
what are the advantages of helping existing suppliers learn to be lean? The strongest
argument against switching suppliers to get new capabilities is that all the benefits associated with long-term supplier relationships might be lost. As Sako (1992) has pointed out, trust between supplier and customer is essential to achieve these benefits, so switching suppliers could hurt not only the relationship with the supplier that lost business but also with other suppliers obseming this event.
In
additio~ the best lean suppliers may have prior commitments to other customers, so they may be less responsive to a newcomer. Finally, the customer has fewer sourcing options if it waits for competitive forces to generate a largerpool of lean suppliers than if it acts to improve the capabilities of its existing suppliers. Steer your supplier to a good consuhant or partner. The assumption here is that a customer should encourage a supplier to develop lean capabilities on its ovq or to seek help tim consultants or
Page5 partners rather than intdkring
directly with the supplier’s internal operations.
Yet the knowledge
underlying lean production is not necessarily easy to transfer across organizations. It seems to require a primarily “hands-on” approach in which key principles are taught by obseming how certain problems are handled in real-life context (MacDuf7ie, 1997). Thus a lean customer may conclude thaG compared with most alternatives, it has superior knowledge about lean production and a greater ability to motivate suppliers to learn.3 Summary. The decision to create lean suppliers is driven by multiple concerns. There cau be substantial diseconornies of vertical integrationoutside of the core business. Switching to lean suppliers may entail considerable costs (economic, political, and reputational).
Helping suppliers become lean
potentially enlarges the pool available for sourcing choices. Finally, customers may be more effkctive than outside parties in teaching suppliers to be lean.
Honda’s Philosophy History.
of Supplier Relations
Honda’s approach to supplier relations is rooted not only in the common Japanese
practice of long-term supplier relationships but also in its own history in the auto industry. Unlike Toyota and Nissaq who were building cars before World War II and developed strong and loyal supplier groups in the postwar era (Nishiguc~
1994; Cusumano, 1985), Honda was founded only in 1948 and began as a
motorcycle company. When Soichiro Honda decided to begin building cars in the early 1960s, he had to develop a supply base from scratch drawing on three sources: 1) suppliers of motorcycle parts, who were already famdiar with Honda but had to learn to make automobile parts; 2) other small suppliers in the surrounding q
who needed to be persuaded to invest in new production capabilities for Hon~
based
on an implicit promise of fixturebusiness; and 3) larger companies supplying other auto companies. Each of these sources of suppliers posed different problems for Honda The motorcycle suppliers who were already in the Hon& “family” were easiest to help, since channels for coordination and technical assistance were aIready established. diliation
The small local suppliers were eager to establish an
with Honda but technologically backward and unftiar
delivery reqired by an exportaiented
with the high reliability in quality and
automaker. The larger suppliers were primarily oriented towards
their dominant customer so Honda had to struggle to get them to be responsive. However, because they
Page6 had more sophisticated technology and superior production expertise, Honda was forced to go to them for cer&ainparts (Authors’ interviews at Honda in the U.S. and Japan; Mair, 1994). There are parallels between these early experiences in Japan and Honda’s early years of developing a supply base in the U.S. Honda’s decision to come to the U.S. reflected a long-term strategy of “make our products where we sell our products” and its corollary for pudmsing
-
“buy our parts
where we make our products.” When Honda started its U.S. manufacturing operations in 1978 (again with motorcycles), it was initially supported by a core of Japanese suppliers already part of the Honda “ftiy”,
a number of which established satellite plants near its Ohio manukturing
complex. Honda
also began trying to identi~ small local suppliers in Ohio and surrounding states to take on certain parts. Many of these suppliers were eager to work with Honda but needed considerable assistance to meet Honda’s cox
quality, and delivexy requirements. Finally, Honda approached some of the larger auto
suppliers whose primary customers were the U.S. “Big Three”.
While these firms had ~P~or
capabilities, they were typically not as responsive to Honda’s requirements as the smaller suppliers. Philosophy. This history helps explain key aspects of Honda’s philosophy of supplier relations. Honda wants suppliers who can be “self-reliant”, with a sufficiently diversified customer base that they will not beat risk if Honda’s orders drop due to demand fluctuations. The importance of self-reliance was a lesson learned from painfid experience during recessions in Jqxuq when Honda’s commitment to its “child” suppliers (small local companies highly dependent on Honda) became an immense financial strain.
Honda also selects suppliers based on whether their management is willing to be responsive to Hon&’s needs. These managerial attitudes are more important to Honda than the suppliefs technical expertise. Examples of the “right” attitude, ilom Honda’s perspective, include 1) a willingness to take risks, consistent with the ‘racing spirit’ that Mr. Honda worked hard to maintain in Honda’s culture’; 2) investments in new technologies in advance of competitors; 3) investments also in organizational and human resource capabilities (e.g. advanced engineering and production control *,
sophisticated
management systems; worker tmining); and 4) doing all of the above without explicit contractual commitments. To a self-sufficient supplier willing to offer this kind of responsiveness, Honda offkred a great deal in returm A supplier to Honda would have a lifktime relationship - a maniage - in all but the most unusual cimmstances.
While specific commitments for fbture business would not be made, suppliers
Page7 could count on receiving at least as much business as in the most recent year past and possibly much more. Furthermore,
due to Honda’s sustained grow@ a supplier who was willing to keep up with
Honda’s stm.tegic direction and production requirements would oflen be asked to take on new parts they had never made before - and even new production processes. It was clear Honda felt it was easier to teach the technical knowledge associated with a different product or process technology than to fmd a technically-capable supplier possessing the combination of risk-taking attitude, motivation to improve, responsiveness to fbture needs, and overall competence that it valued so highly. The importance of understanding Honda’s supplier development activities in the context of this broad vision of mutual responsibility and obligation between supplier and customer cannot be overstated. In the words of Rick Mayo, the Honda engineer directing these activities, “We are a philosophy-driven company. We do supplier development as a way to teach our philosophy, to put it into action. It’s how we tIY to help suppliers get past their old way of thinking and understand our way of thinking. It’s a missio% not a job.”
“BP” at Honda BP is the core supplier development activity at Honda of Anerica Manufacturing (HAM). HAM has a Supplier Development Group with 50 staff members in its purchasing depatnmnt that oversees BP and other supplier improvement activities (Celeste and Sabety, 1994). Once a supplier is chosen for BP (see below for more discussion), a few staiT members from this group along with employees fivm other Honda departments (e.g. vehicle quality, process engineering) form a BP team with supplier employees to work for several weeks at the supplier’s f=ili~. work areas, and hitidly
The BP team focuses on improvements at a few specific
avoids projects that would require extensive capital investment (“hard” BP) or
extra personneL In@e@ BP tries to cover all aspects of a narrowlydefined project - technology, work organization
problems with second tier suppliers, or woridorce issues (e.g. motivatio~
tmining,
compensatior4 employment securit@. The narrowscope allows quick results, which provides motivation for BP participants and data to convince skeptical managers to continue backing the effort. BP’s deep analysis (only feasible for a narrowlydeflned project) helps teach systemic thinkbg, which can then be applied to other areas within the supplier’s plant. For the lines on which the BP team focuses, perfommnce improvements are large
Page8 Honda reported productivity increases averaging 50’%0at the 53 Honda suppliers participating in BP as of 1994, and seven ilrms interviewed for a report on BP reported productivity gains of 25% and quality gains of 66’%(Celeste and Sabety, 1994, p.34). The goals of BP are consistent with Honda’s production philosophy for its internal operations: 1. Encourage fksh thoughts about production processes 2. Gather better data to allow for more thorou~
fact-based problem analysis
3. Seek “common sense”, low cost solutions by following the “five why’s” 4. Know the context by examining the “actual part, actual place, actual situation” (the 3A’s) 5. Create a smooth flow of production with no waste. The firsl BP goal of “encouraging fresh thoughts” acknowledges the need to shake an organization out of its routine ways of looking at its production process or a particular problem.
Each BP team
contains members from various departments and levels in the supplier’s organization to insure varied perspectives.
The second goal of gathering extensive data also helps with breakhg away from past
routines, particularly since Honda has found that many suppliers don’t keep records about their processes in ways which make it possible to determine if a change leads to improved performance or not. The third goal involves “root cause” analysis. Here Honda teaches the process of asking “why” five times, established by Toyota’s Tai.ichi 0hno.5 To do an effective “root cause” analysis requires considerable contextual knowledge. The fourth goal of BP is to develop that knowledge by going to see the “actual part in the actual place and in the actual situation.”
Honda’s BP representatives try to
demonstrate this principle at all times, insisting that BP team members go to the actual spot on the shop floor whenever a problem needs to be explored-not sit in their offices and analyze the problem abstractly. Finally, the fifth goal of BP is the elimination of “muds” or waste wherever possible. Waste is defined as anything that interferes with the smooth flow of production.
For example, if tools or
components in a given work area don’t have a specified storage place, workers may have to hunt for them. Even if this takes only a f-
seconds, the time wasted can be substantial if multiplied across weeks and
months of production activity. The effort to achieve a smooth production flow reveaIs msny upstream and downstream problems, e.g. with other operations in the plant, with suppliers, with the customer order process, or with the distribution system. When each of these is pursued to its “root cause” and remedid huge amounts of “muds” can be eliminated.
Honda required only a few things iiom the supplier
Page9 companies participating in BP. A supplier did not have to pay for the time of Honda’s BP team members but was responsible for the cost of tools and materials requiring for improving core production processes; as noted above, for most BP projects, these latter costs were minimal. The supplier also had to agree not to carry out any employee layofi tiormation
as a result of BP activities.
Finally, Honda required ready access to
about a supplier’s cost stmctum and technology, the ability to move about freely within the
supplier’s production fmility, and the cooperation of management in efforts to involve fkont-line employees in BP improvement projects.
BP Case Studies We turn now to case studies of BP projects at three of the suppliers we visited b-een
1992 and
1994. We pay particular attention to the overall relationship between the supplier and Hond% because it so greatly influenced the BP process.
Capitol Plastics Capitol was one of Honda’s earliest U.S.-owned suppliers, going back to the late 1970s~ Faced with declining business for the Big Three, Capitol began exploring opportunities to work with Japanese customers.
After four years of contacts, Honda made them the second supplier of a small motorcycle
fender part in 1978, in order to obsme their capabilities. The owner, who was also managing the plant at the time of our 1992
vis& recalled
how much information Honda asked for in the early days of their
relationship: Other suppliers thought I was crazy to give them so much. But we found out that the dialogue was genuine. First we got an order, then a purchasing agreement. Then our volume began to expand. All very simple, very little written dowm When Honda began to build cars in 1982, they asked Capitol to take on some important console parts for the Accord. At ~
the tooling
was
supplied from Japan. Honda also arranged for Capitol to
establish a “technical collaboration agreement” with their supplier of the same pm in Japan. This firrq Morioko, began to send technical personnel to Capitol to help with equipment hallation launch
and product
Page 10 Keepingup with Honda’s dramatic growth over the next 5-7 years was like “trying to keep hold of a tiger”, according to the owner. “we showed we’d do whatever they wanted.” Capitol’s overall sales grew from $7 million in 1979 to $30 million in 1989. But theu in the industry downtuxn of the early 90s, Capitol lost a big chunk of business for a Big Three customer due to a design change that eliminated their
part. In the resulting crisis, Capitol went to Honda to find out what improvements they would need to make to continue increasing their share of Honda parts. Honda said that Capitol should get involved in BP. Despite Capitol’s willingness to be responsive, Honda was tied
with Capitol’s inability to
resolve persistent quality and delivery problems. They sent a team of 4 people from the BP group to “live” at Capitol for 9 months. Two had engineering backgrounds and two were former assembly line workers. This BP team focused first on changes at the management level, recommending changes in reporting relationships and redeployment of certain managers in order to get the more people-oriented managers into shop floor positions. Then they undertook three BP projects in the plant, each devoted to a single production line and overseen by a task force composed of opemtors, engineers, and Honda’s BP representatives. ‘Ihe first two projects took 3 1/2 months each and the third took 2 months. The emphasis in each project was two-fold: how to evaluate the line and plan improvements, and how to get workers involved in the process. Improvements to the line included the redesign of machine layout to reduce walk time and other unnecessary motio~ converting parts racks to a “flow” design (in which gravity pulls a new bin of parts down a slanted shelf with rollers once the old bin is removed), constructing circular fixture to hold parts (height-adjustable for diffkrent workers) that can be loaded on one side, spun around for work to be done, and spun again to be unloaded; improving working conditions (e.g. rubber mats on the floor where people stana better lighting immediately over the work area), and replacing hand-held tools with fixed-position tools on moveable trollies for better process consistency and less strain on wrists and arms.
During one of the projects, a task force made up of workers fbm the BP line, organized by Honda’s BP st@ uncovered the “root cause” of a mysterious qyality problem. On an irreguku basis, parts would emerge from the molding machines with “splay” (white spots along the edge of the product) or “short-shots” (a mold not completely filled in). The workers discovered that these conditions resulted from W4 cold particles of plastic resiq which were, in ~
caused by condensation falling into the resin
Page 11 container from an exhaust fm in the roof
Once diagnosed correctly, this was an easy problem to fix.
Their success inspired great enthusiasm for BP among the task force members. Honda believed it was crucial to teach supemisors and engineers at their suppliers how to be responsive to operator ideas - a diilicult cultural change. As one Capitol employee put it: Their ~onda’s] philosophy is to pull the lower skilled people into the job of redesigning the line. They’re willing to lose some on the technical side. They’re not just after the best process but getting the people motivated getting their skills involved. This employee then expressed reservations about how well this approach was working at Capitol: I don’t know that Honda’s approach is the most cost-effective short-term. At Hon~ the processes are already set and pretty goo~ so when workers make suggestions, the changes are small. Here and at most small American suppliers, the process is ~ well-established. You need to get the process fixed first, and that’s a ~ change, which means a lot of fear. 70% of the people would like to participate but 40% are afraid to participate. Sometimes the whole process feels too slow - but I’m not sure management or supervisors could do my better. The three BP projects brought big improvements in throughput time, inventory levels, scrap, quality defixts, cleanliness of the work are% and injuries. For example, the BP project for a part called the Box Instrument Back produced a 45% increase in productivity and a 67’%0reduction in the scrap rate. For one BP proj~
some subcontracted work was brought back to Capitol for more control over quality and
to maintain employment levels as the line grew more efficient. However, tier
the Honda BP team left, problems crept back in.
Delivery problems were
particularly severe due to Capitol’s lack of expertise in production planning. Capitol also failed to capitalize on the enthusiasm unleashed by the initial implementation of BP. Two examples were cited by both managers and union officials during our visit. in the first case, the BP team for one line suggested covering the areas where workers stood with rubber floor mats. Though they cost only a few hundred dollars, the mats proved effective in reducing operators’ fatigue from standing on a cement floor. Other workers saw the mats and asked if they could be installed on their lines too. However, they were told that any changes had to wait until BP came to their line-a time which was suby.ct to many delays. In the seared case, workers on another line could not find out how much money their improvement efforts had saved the company. Disclosing this figure (an amount which turned out to be $250,000) was a standard part of the Honda methodology and a key motivator for participants, who wanted to know that their extra
Page 12 effort had produced something worthwhile. However, the owner did not want to disclose financial information to workers, because he feared that the union would use it to win wage increases. Capitol’s diflicuky in keeping the BP process going and the persistence of quality and delivexy problems concerned Hond~ as well as the firm’s owner. Under guidance from Honda’s BP advisors, managerial assignments were ~
new staff were hired, and new intensive projects (e.g. a
delivery improvement project with another stti group at Honda) were undertaken. Honda’s investment in helping Capitol improve was now quite high; Capitol managers estimated that Honda’s technical assistance had cost the automaker over $1 million.
But it remained unclear whether Capitol had the
resources for the investments in both physical and human capital that would be necessary to meet Honda’s continuously increasing demands. By the time of our visit, Capitol was actively exploring purchase offers and several months later, it was purchased by a large, diversified company making many dif%ent kinds of parts.
Tower Automotive This firm has its roots in Western Michigaq an area which was settled by Dutch immigrants in the early 19th century and has strong religious traditions that are grounded in the Dutch Reform church. The area is known for a strong work ethic and progressive employers. Social ties outside of work are common among managers and employees. Firms have long been innovators in employee involvement and gain-sharing plans and they typically make a strong commitment to avoiding Iayofik Honda purchased parts from several firms in this ~
we visited both Tower Automotive’ and Donnelly Co~oration*.
Honda found Tower congenkd born the start. Its older plants were crowded and cozy, reminding some Japanese managers at Honda of their older plants in Japan. The long-serving, loyal workforce was another characteristic Honda hoped to encourage in all its suppliers. Yet despite this codort
leve~ Honda
wanted much that was difficult for Tower and its counterparts to provide. Tower was representative of a group of technically compete@ dependable, long-time suppliers to the U.S. Big Three automakers, accustomed to the “boom or bust” cycles of the industry and to keeping costs low to avoid being undercut on price by mnpetitors.
Fundamentally mass producers of commodity products, these suppliers exploited
economies of scale while also relying on unique technical enhancements to achieve price and performance advantages over competitors. These firms were not, when they began working with Hon4
particularly
Page 13 ffiar
with lean production.
Insteti
they were successfid enough as mass producers to have little
reason to change their approach Honda learned about Tower Automotive in the late 1970s, when it sent a team to visit over 100 U.S. stamping firms listed in the directoxy of the Precision Metal Forming Association. For Tower, which was accustomed to seeing a Big Three buyer only once every five years, the f~ that they were visited by the president and members of the board of Honda of America made a strong impression. One Tower manager in particular became a strong internal advocate for the advantages of working closely with Honda. Towe/s experience with Honda moved through a set of small, steady steps, each of which impressed Tower’s initially skeptical management. worked extremely well from the stat.
First, the tools that Honda provided from Japan
Then Honda provided some steel from a Japanese supplier that
proved very easy to work with. Eventually the steel Tower received from Inland Steel - a major U.S. supplier to Honda - began to show the same attributes as Honda worked with Inland on process improvements. While Honda and Tower had very different attitude towards tool design and maintenance, the two companies were able to work out agreements on acceptable modifications in tooling relatively quickly. Satisfied by these early experiences, Honda began to give Tower more business. As volume grew, Honda asked Tower to build a dedicated plant for Honda Tower refbsed at first but in 1987 did agree to put Honda production into a pl~
in Aubuq
Indian% that was initially built for Ford in 1985. Tower
then expanded the Auburn plant four times in the next few years to accommodate the higher volume for Honda Tower managers were not always comfortable with Honda’s policy of having suppliers provide lots of detailed information about their plans and operations. With BP, at first there was a lot of nervous tension. We didn’t want to give away the farm. Eventually we redid that Honda’s wish to know everything is not because they are trying to steal our good ideas or because they want to be snoopy. They want a partnership and want to be able to help you find the best way to do things. Honda’s zeal for results, attention to detail and wish for involvement on the shop floor stood in contrast to ForA the other primary customer at the Auburn plant. Honda’s BP people visited Auburn twice a week
In additiou quality control people came from Honda twice a month and Tower people went to
Page 14 Honda in Ohio twice a month. As one manager told us, with Ford you “never” see the buyer in the plantjust the quality co-l
people once a month. “We go to Ford as little as possible - it is our plant’s turn
once a quarter.” The flip side of Honda’s attention was the strong pressure it applied when problems occurred. Tower (like other Honda suppliers) gets a written report any time there is a quality or delivexy problem and has to give a written response detailing what “countermeasures” it will take to remedy the problem. Hon& keeps track of the number of parts that are def-ive
or delivered late. If a shipment is one part
short, the entire number of parts in that shipment are counted as defective. As one manager explaine~ Ford in contri@ is more likely to “call us up” when they see a problem and even then only if it om.m often enough to appear to be a pattern. If a shipment is a few parts short, they’ll tell us to “throw a few more” in the next shipment. Tower managers differed on the merits and shortcomings of Honda’s and Ford’s approach. One manager preferred Ford: Ford has focused on systems. They believe that if you have good quality control systems in place, you’ll have good parts. After the systems are in place, they leave you alone as long as you’re performing. With desigq we’re more involved at an earlier stage with Ford. They give us CAD (computer-aided design) data as the master specification and then let us work from there. When there are problems, they handle them infomdly, over the phone. They get a quicker response that way than if they write us up and “ding” us. This manager also felt that Honda had benefited from Towefs adherence to Ford’s SPC system: Ford really dragged their supplier base up. Ford was the fimt to teach us SPC. 75% of our salaried people went through their program at the American Supplier Quality Institute. When Honda saw we had Ford’s system, they urged us to use it. Another manager found value in Honda’s obsession with eliminating clef-: Honda cares about making the part fit the car, while Ford cares about making the part fit the bhwprint. During product Iauncb Honda takes parts as soon as they are made and runs back to try them on the car. Then they tell us to change this, change that. Ford usually isn’t here during our trials. They just want to be sure that we are meeting the spec. If there is a probl~ they eventually issue an engineering change. But at Hen@ the changes happen in a matter of days. At first we thought they were nuts. But theirs is a great way to do business. You get what you want - a part that works on the vehicle right away. Everythhg else - like whether the blueprint is up-to-date - is secondary. Initially, it was incredibly fhstrating because Honda was so detail-oriented and wanted responses finm us immediately. But I find they are almost always right.
Page 15 While we were in the planq we saw an example of how Tower combined Ford’s emphasis on procedures with Honda’s focus on an attitude of continuous improvement. We had asked to see a statistical process control chart, a type of documentation required by Ford but not by Honda for this product. The manager accompanying us noticed that the process was heading out of control on a key dirnensiorq he and the operator quickly engaged in a discussion of what might be causing the problem and how to stop it. BP started at Tower in late 1989 with the arrival of a team of three people born Honda - born welding, quality, and purchasing departments. The team proposed starting with “soft” BP (simple projects with low investment) - organizing all steel blanks in one are% painting floor spaces to indicate where steel coils are to be plac@ attaching new bins to stamping machines that are more accessible and can be repositioned. Next came a project involving “hard” BP (i.e. investment in new equipment) - reengineering the work cell for the center pillar (i.e. a stamped part located between the flont and rear passenger doors that connects the roof to the side panels). Initially, Tower’s work cell utilized dedicated automatio~ designed for a given center pillar and able to produce 63 parts per hour. Honda encouraged Tower to invest in robotic technolo~,
but of a particular sort. Rather than having the robots move around the part to apply
welds, the welding gun was on a pedestal with a fixed location and welding position. Simple, low-cost, and reliable “pick and place” robots could then be used to move the parts around so each weld could be applied. Automatic sensors were used to check whether all welds were completed successfidly. With this new cell desigq productivity rose to 125 parts per hour.
The life of weld tips was also increased
dramatically, from 50,000 welds before tip replacement to 250,000 welds, because the fixed position of the weld gun meant less wear and tear on the tips. manuktming
Over time, Tower was able to influence Honda’s
approach. In the words of a manager at another supplier, as Honda came to trust a firm’s
technical capabiliti~
a supplier could “cam the right to disagree.” For example, when punching holes in
a stamping, Tower used a “button” of metal underneath the part that the punch would strike, hsuring that the punch would always go to the same depth Honda was accustomed to doing such a punch without a button underneath - something that increased the risk of splintering around the hole and also required more maintenance for the punch tool. Tower persuaded Honda to install easily-changeable buttons and to make them of hardened steel to reduce maintenance.
Page 16 Despite acknowledging learning a lot horn Hond% Tower managers and engineers still resisted Honda’s pressure to provide lots of idormation. When Tower built its Auburq Indiana plaug the first that Honda heard of the plan was when the supplier presented Honda with blueprints. While Honda was pleased that Tower was building a new plant, they would have liked more opportunity to comment - a desire that Tower felt was intrusive. In another case, Tower felt that they had withstood pressure from Honda and had decided to go their own way. At its Greenville Michigan plant, Tower hired a consultant to implement a “workplace transformation” program that dii%ed from BP. Whereas Tower managers saw BP as focusing on value analysis and process reengineering, they saw the new program - whick once adopted at Greenville, replaced BP there -as
focused on the implementation of independent business units within the plant and
“self-directed work teams” on the shop floor. The move to self-ed
teams also had an explicit role for
the union (the Greenville plant was organized by the UAW, whereas the Auburn plant was nonunion) and emphasized efforts to improve job design. While Tower felt that Honda disapproved of this initiative, Honda disagreed. As one Honda manager explaine~ “Our first question whenever a supplier proposes something is ‘why’? We just want to see if it makes sense. We find that if the supplier doesn’t have a good answer, it’s a sign of potential problems. But the supplier reaction is often ‘Honda doedt like it’.” What struck us about this incident was how unfarnilk and uncomfortable the suppliers were with the idea of talking over their internal plans with a key customer. Clearly suppliers are most accustomed to a situation in which a meeting with your customer is a time for receiving orders. Honda has found that to encourage “f&h
thoughts” iiom suppliem, they must be quite flexible about the extent to which
improvement activities fd under the BP rubric. A&r the workplace transformation program was launched at Greenville, efforts began to bring it to the Auburn plant as well. This took the Auburn plant manager away fkom the plant in order to help train other managera, right at the time of the 1994 Accord launch
Honda was concerned that its
implementation at the same time as a major product launch was drawing attention away flom persistent qw@ problems with 1994 liccord p-. Tower also asserted its own wishes in some investment decisions. The quality problems alluded to above were concentrated in parts for the Accord station wagon. Because the volume for this product is relatively low, Honda and Tower agreed that investment costs should be kept low and that dedicated welders should not be installed. However, when problems persist~
Honda pressed Tower to invest in
Page 17 automatic sensors that check for missing welds or nuts on subassemblies. Tower refhse~ insisting on all-manual processes.
In another case, Tower insisted on automating processes that were pdormed
manually at suppliers making the same parts for Honda in JapaIL despite Honda’s concerns about the impact on overall costs. One factor affkcting Tower’s relationship with Honda in recent years is its acquisition by Hidden Creek Industries and its consolidation with sevemi other medium-sized stamping compsnies in the area Tower Automotive, the consolidated firm, has made a determined bid to absorb lots of new business horn FoI@ Toyo@ and Nissan. With Ford reducing its stamping suppliers born 400 to under 50, Tower was determined to stay in that select group. respo~ibili~.
Ford was also increasingly willing to give Tower major design
Furthermore, Tower was building a new plant in Kentucky to seine Toyota and Nissan.
Thus while the Auburn plant remained dedicated to Honda (as it had been since 1990), managerial time and effort for the newly-consolidated firm was pulled in many different directions. So despite a strong and long-lasting relationship, Tower has struggled with Honda in many ways. Honda continues to press Tower for performance gains and for continued investments in improved capabilities.
Faced with pressures from two major customers - Ford and Hon& - ToweI’s senior
managers seemed to be finding Ford’s more “hands+ff
approach more appealing. With Ford’s emphasis
on getting good systems in place, Tower was able to satis~ Ford’s expectations on a regular basis, while Honda’s insistence on responsiveness for each new defect or delivery problem was more annoying. (As one manager explaina
“Ford’s sub Iiminal message is that a few defects are part of life - get your systems
right and you’ll basically be OK. Honda is after us with their Supplier Performance Report every month”) Also, Ford seemed more likely to make defmite commitments of volume to Tower in advance of investment decisions, while Honda continued to insist that their “partners” continue to make needed investments so they would be ready for whatever new business might be available. Thus the Tower case suggests that even where BP efforts succeed in boosting a supplieis technical skills, capabilities for improvement and overall performance, the stress of responding to two major customers with diflhnt tight like.
Progressive Industries
priorities can still prevent a supplier from being as responsive as Honda
Page 18 Progressive is a small stsmping fkm that had a long history of second-tier contracts with the Big Three before it began to work for Honda It is a family-run business now headed by Ruston SimoU son of the founder!
His father was an experienced tool-and-die engineer, but Ruston went to college at
Northwest~
worked a few years in banking in Bosto% then got an MBA before taking over the family
business.
This combination of education and experience gave him a nontraditional
(and more
sophisticated) perspective on how a small auto parts firm should be managed. Progressive was founded 40 years ago as a “die shop” that built tools for bigger stamping companies. To test the tools they were building, they bought an old stamping press. Then a customer asked them to take on the stamping of a small part. They found the steady work associated with an auto contract to be more appealing than the “feast or ftie”
cycle of the tool and die industry, and added more
and more presses. Many small stampers - tagged by some in the industry as the “smash and ship” bunch - had a similar histoxy. Keeping costs low was paramount, so physical plant was kept small and sp~ wages were low, presses were old and high capital utilization was critical. Companies like Progressive that were able to use their tool-and-die skills to keep machines running had some slight advantage over other commodity stampers, but price competition was fierce and customers were willing to switch suppliers with each new low bid. When Simon took over Mogressive in 1984, the company had moved away from the automotive business to avoid its cyclicality and price competition but was not particularly profitable. He decided to move back into the auto industry, but discovered Progressive had f~en pdormance
considerably behind the
levels needed to win contracts. He began approaching other small stampers in the area who
were f=ing capacity cmmtmints, offering to make their most difficult part for them. Progressive was able to help these firms improve the pdonnance
In additio~
of their tools.
When Progressive decided to pursue major auto contracts directly, it found the Big Three were shedding suppliers speedily.
Then in the f~ of 1986, Simon spotted a Honda ad in a metalworking
magazine, with the headline “fVANTED: competitive Stamping Sourca “(see Figure 1). Honda sent an extensive questionnaire asking about the firm’s history, equipment, profits, “ever@ing”.
Ther4 as it had
in the late 1970s (when it found Tower), Honda sent a team of 3-4 American and Japanese purchasing associates to visit 120 stampers in a five-state area
Progressive was one of seven stampers chosen
through this process, and the smallest. Honda was impressed with Progressive’s quality systems and even
Page 19 more so with the rnauagement team’s openness to using Japanese tools and learning Honda production methods. Progressive’s experience with Japanese tools is instinctive, because the design philosophy for dies differs dramatically between the U.S. and Japan. U.S. dies are typically engineered with very hard steel, in anticipation of a long life cycle for a given car model, and hence are both expensive to buy and need a long lead time to build
Japanese dies are typically cheaper, made of soiler metal, and require a shorter
lead time to develop; however, they also need fkquent maintenance. This fits well with the shorter life cycle for most Japanese-designed autos, since extra maintenance costs are easily offset by the quicker development process and cheaper raw materials. By using Japanese tools, Progressive’s tool and die makers gained exposure to this very different philosophy of die design and maintenance. Secont they developed idem about potential improvements in the Japanese dies an~ after producing successfi.dly with these tools for a few years, persuaded Honda to switch to U.S.designed dies incorporating these ideas. Thus Progressive was able to show that they were receptive to new ideas but also able to contribute their own innovations. Progressive was particuhrly innovative in the use of die-sensoring - mechanical and electronic gauges on the surfaces of dies and edges of presses that monitor whether the stamping equipment is operating within certain critical parameters. For many traditional stampers, this monitoring process was done by assessing the sound of the machine and the shininess of the areas where die surfhces meet. With Progressive’s home-grown sensors (and the accompanying sofhvare, developed with outside vendors), presses could be calibrated more precisely, to use the “minimum pressure required, which reduces weq saves energy, and allows more precise matching of jobs to the tonnage of the equipment. Mso, sensors can quickly shut down machines as soon as they detect a problem, similar to the “jidoka” philosophy associated with the Toyota Production System. The information generated by die sensoring became important in one of the BP efforts Honda carried out at Progmsive: the development of very detailed technical standards and operating procedures for each press, to systematize production processes, limit variance that could lead to clef-,
and allow the
accumulation of process knowledge that could lead to further innovations. While Honda was relatively satisfied with Progressive’s technical capabilities as a stamper, it pushed the firm to move to the next “value-added” step of welding and assembly of stamped parts to produce a finished sub-assembly.
Unlike U.S. automakers, who were more likely to build such
Page20 sub-assemblies in their own plants, Honda has long preferred to give responsibility for sub-assemblies to suppliers. At Honda’s small stamping suppliers in Japa
roughly 2/3 of their workforce is devoted to
welding and assembly tasks. These were new areas for Progressive, but adding these capabilities was appealing to Simo% who wanted to move the company away from being viewed as a mere “commodity” producer. “We needed to eam our spurs with them” on welding and assembly, he admitte~ “and we were given more leeway at first than the bigger guys. They started us off on small, simple tasks and we were able to show good results early.” Increasingly Honda is pushing Progressive to take on more complex parts and weld patterns, moving towards multiple welds along multiple axes. Progressive’s tool shop heritage has been applied to in-house development of customized welding equipment for the Honda parts they make, including robotic cells. The BP projects at Progressive unfolded against this backdrop of increasing responsibility, technical challenge, and expansion into new processes. According to Progressive managers, “there was no single, critical event that was plant+hanging.
It was a steady learning process. Honda brought us along
very slowly.” One BP project developed a color-coded priority system for die maintenance. /mother BP project took a single press and made a series of small improvements: 1. The control panel was moved tim the side of the press and put onto a rotating arm, so it could be moved close to the operator while he/she observed press operation. 2. Small welded compwtrnents to hold tools during maintenance were attached to the press. 3. Oil needed for press maintenance was piped from a central storage locatiom rather than being hauled in bands and poured into line-side storage containers. 4. A staging area for steel coils was created upstream from the presses, to speed changeovers. 5. Simple md dividers were installed to insure that scrap pieces would fdl automaticdy into scrap containers without having to be handled. 6. Three bins for finished parts were put onto a rotating stanL so one bin could be filling while another bin was being unloaded into shipping containers. 7. The floor area around the press where incoming materials or outgoing parts are stored was paintedto indicate precise placement locations for each item.
Page21 Complementing Progressive’s BP efforts on the shop floor was a human resource system that was consistent with Honda’s values an~ in many cases, influenced by Honda’s policies. Progressive has had only one layoff in its 40 year history, so the firm had no problem accepting Honda’s no-layoff condition for BP. To absorb volume fluctuations, Progressive has established a work schedule of 45 hours (4 10-hour days and 5 houm of scheduled ovtiirne), so that it could cut back to 40 hours when times were lean or add overtime in boom periods. Besides their relatively low base pay ($8 for operators and $15 for skilled trades), employees received profit-sharing based on meeting targets in each quarter, plus a discretionary bonus ($980 in 1993) at year’s end. Progressive also provided W reimbursement for outside education and has boosted its training significantly, often using training materials developed by Honda Like Capitol Progressive’s euly responsiveness to Honda brought a steady growth in orders sales to Honda that more than doubled from 1992 to 1995 and an increase in first-tier and second-tier parts produced fiwm 10-20 in the late 80s to 125-130 by 1994. Unlike Capitol, Progressive has been able to manage this growth effectively while also taking on additional responsibilities.
More money for
investments, a more technically sophisticated staff, and more attention to human resource development all appear to be part of Progressive’s success. The Progressive case is a good example of Honda’s preference to give additional business to suppliers that have demonstrated responsiveness, good performance, and the wdlingness to take on new challenges.
Analyzing BP’s Effectiveness
and Supplier Capabilities
We start our analysis of the supplier cases by evaluating the effectiveness of BP. (See Table 2.) FM
we assess the scope and mstahmbility of the organizational learning associated with BP at each
supplier and the absorption of the krzizenphilosophy of continuous improvement.
For all three suppliers,
the adoption of a kuizen philosophy represented a significant change in their organizational culture. For a small company like Regressive, whose owner became a quick adherent this culture change occurred relatively quickly. Small size did not help at Capitol, where control-oriented management thinking and limited staff capabilities hindered the prospects for katien. The move towards kaizen was slower at Tower than at Progressive, yet there were more resources to sustain change once it occurred. Next, we consider the extent to which each supplier took on responsibilities and developed capabilities for being an effective Iong-term partner for Honda. Of the three suppliers, both Tower and
Page22 Progressive were quite effive
in meeting Honda’s basic cost, quality, and delivery requirements; only
Capitol was weak in these areas. In additio~ Tower and Progressive had “earned the right to d.is~e” with Honda with respect to mandhcmring processes, ptiicularly
the modification of die design and
maintenance practices. AU three suppliers grew dramatically due to Honda’s business. But there was more variation in the extent to which these suppliers made major investments in new technologies and new facilities in order to keep up with Honda’s fbture requirements. Tower, a relatively large supplier (and larger now since its consolidation under Hidden Creek), made major investments, while the smaller suppliers did not - primarily a Iimction of the greater ability of a larger firm to gain access to capital. Progressive did add subassembly tasks in order to become a more %Jl service” provider to Honda On balance, the cases support the idea that creating lean suppliers is a viable alternative for a customer that doesn’t want to vertically integrate; does not have easy access to a more highly skilled set of suppliers; and wants to maintain commitments to existing suppliers. But the cases also reveal that this approach is not easy. Next, we look at the evidence presented above, together with observations horn ●
recent interviews with Honda managers, and combine them with insights horn theories of organizational learning to suggest generai lessons for customers and suppliers about when and how best to undertake knowledge trsnsfm processes such as BP.
Lessons For Customers Structure the learning process so that the knowiedge is easier to absorb.
Organizational
learning and innovation theorists argue that technical knowledge is particuhrly difficult to transfw if it possesses the following characteristics: an abstract or complex scientific base; a ‘%agile” technology that doesn’t work consistently; technology that requires “hand-holding” because of idiosyncrasies that make standankation
impossible or “lumpiness” because knowledge transfm would affkct lots of people at the
same time (Eveland and To-,
1980); a ‘tacit’ or ‘unmodified’nature that is difficult to explain
explicitly (Nelson and Winter, 1982); a ‘radical’ or ‘competencedestroying’ nature that makes obsolete some of the recipient’s pm+xking
capabilities (Abernathy, Clarlq and Ki@fOW,1983).
Honda (consciously or not) designed the BP process to avoid most of the characteristics of technical knowledge that make it diflicult to transfm. F-
the underlying scietic
lmowledge for the
reengineering of production lines was primarily concrete and simple rather than abstract and complex. BP
Page23 reinforced this by following the principle of “actuai pat, actual place, actual situation” (the 3A’s) and by fmusing attention on a single problem. Second, the solutions developed through BP were highly reliable. These solutions routinely produced fhst, large improvements in suppliers’ operations-reinforcing
the
incentive to continue with the program. Third, standardization of processes was ofien achievable and was encouraged as part of a continuous improvement cycle that alternated periods of standardization and periods of experimentation with improved methods. Four@ by beginning BP on a single “pilot” line and then slowly diflbsing the same principles to other redesign efforts, the “lumpiness” of broad sweeping organizational change - and the potential organizational resistance to such change - was minimized. Furthermore, BP was well-suited to the tacit nature of the knowledge underlying lean production and continuous
improvement.
In addition to practical illustrations of how to reengineer a discrete
production line, the entire BP process modeled the behaviors that Honda hoped the suppliers would adopt - utdizing the “3 A’s”, gathering “before and after” data to evaluate countermeasures, emphasizing minimal capital investment until process steps are organized for smooth flow, respecting worker Imowledge as a source of ideas for process improvement, and paying attention to (non-wage) factors that 9
dissipate worker motivation. Though we judge BP to be an effkctive mechanism for transferring technical (and often tacit) knowledge, it is worth noting two other characteristics of lean production that affixt the knowledge transfer process.
On the one han~ lean production makes obsolete many of the skills, attitudes, and
heuristics developed under mass production (e.g. emphasizing economies of scale and production targets over quality, establishing buffers to keep production running despite defective parts, high turnover, and poorly-motivated employees). It requires alternative perspectives on many dimensions of production (e.g. accepting a “quality first” philosophy, understanding the logic of but% reductiom becoming more open to employee input into production problem-solving) that, taken as a whole, represent a radical shift in managerial rnindse& On the other Iuu@ lean production is not entirely “competencedestroying”. Indee~ successfid implementation of lean production places heavy demands on traditional skills such as production schechd@
workflow planning and data management. Part of the challenge in learning lean
production is understanding which aspects of traditional manufmg
practice provide a necessaxy
foundation and which represent barriers to adoption. Choose the knowledge reapient carefully, keeping in mind how the recipient’s “absorptive capacity” and “identity”
will affect the knowledge transfer process.
Honda’s choice of BP
Page24 participants was based on a variety of fkctom, some of which heavily affkcted the Imowledge transfti process. All participants were expected to meet the following criteria highly motivated to le~
willing
to make their operations completely accessible, and willing to commit to the “no layoff policy that Honda saw as crucial to BP success. Beyond those criteri% there were two predominant paths leading to being selected for BP.
One path (Capitol) was to have performance problems that persisted despite
Hon&’s standard feedback reports and pressures for improvement. Another path (Tower and Progressive) was to demonstrate strong capabilities in about learning ~
areas valued by Honda and an attitude of responsiveness
capabilities to meet Honda’s current and fiture needs. Honda’s experience with the
suppliers selected via these two paths differed considerably. Two concepts - absorptive capacity and organizational identity - are helpfid as a way to characterize what Honda experienced. One key fmor in the successful transfm of knowledge, according to Cohen and Levinthal (1990), is the “absorptive capacity” of the recipient. The capacity to absorb new Imowledge depends critically on the level of related knowledge that already exists in the recipient ~ since prior knowledge facilitates the assimilation and exploitation of new knowledge. While ultimately grounded in individual
learning, absorptive capacity at the organizational level rquires
effective
mmmunication mechanisms between members of the firm and sources of knowledge inside and outside the firm. The absorptive capacity of a firm is also related to its organizational identity. As characterized by Kogut and Zander (1996), the “identity” of the firm is defined by the organizational boundaries which indicate who is (and is not) a member of the organization by shared goals and values, and by patterns of interaction among individuals that give rise to a common language and common frameworks for actiom A firm’s ability to absorb technical knowledge will be greatest when both the new knowledge and the firm’s prior related knowledge are close to the core of the firm’s identity. In con-a
strong identity
will create obstacles to the absorption of radical change, because the firm’s identity is typically wrapped up in a stock of previouslydcveloped
knowledge and organizational routines
These two factors (absorptive capacity and identity) have different implications for the process of lmowledge transkr at large vs. d
suppliem. Larger suppliers should have more absorptive capacity
than smaller suppliers, both because of more prior related knowledge of the traditional skills on which lean production draws, and because of a stronger identity, based on particular areas of technical expertise, long histories of successfid pdormance,
cohesive corporate cultures and high employee tenure.
Page25 However, this strong
identity
might also increase the resistance of larger .suppkrs
“competence-destroying” change of mindset associated with adopting lean production.
to the
In contrast,
smaller suppliers should have less prior related knowledge about lean productio~ and thus less absorptive capacity. While this might make the knowledge tmnsfer process longer and more time-intensive, a small supplier might be more responsive to the customer’s suggestions than larger suppliers. There can be tremendous appeal for a customer trying to “create lean suppliers” in building up smail firms into intensely loyal medium-sized ones. Their motivation is high and they are likely to mold their “identity” very closely upon yours. However, there is a risk in this strategy of too much dependence and too few resources for developing advanced capabilities. This risk is apparent in the Capitol Plastics case.
Capitol’s strenuous efforts to be responsive to Honda led at first to a larger and larger share of
Hon&’s business.
When the limitations of Capitol’s management team, production expertise, and
technology became clear, Honda used BP to try and bolster the company, at considerable effort and expense, but to little avail. The akernative strategy of working with already established firms that have a strong customer base but still have ample room for improvement (such as Tower) seems more promising. The early stages can be considerably more compleq since the identity struggles are fierce at the beginning. Picking the right supplier-partner can mike a big difference during this stage - e.g. the initial “fit” in company cultures between Honda and Western Michigan suppliers such as Tower, even if far finm tight, was sufficient to provide a strong core to their relationship with Honda even during difficult times. Certainly the fact that the fundamental absorptive capacity is present means that the “provider” firm can concen~e energy on motivating the recipient firm and minimizing the “stickiness” of the knowledge transfm process. However, even once the initial struggles are pm
the responsiveness of larger firms may be
limited by the competing demands they face from diffkrent customers, as was the case at Tower vis-a-vis Honda and ForrL HowevcF, the case of Progressive Industries suggests that it is too simple to assume that large suppliers are more capable of karning fhm a lean customer like Honrk
BP was largely successfid at
Progressive due in part to the “absorptive capacity” of a key individual - the firm’s owner - who saw the need to keep expanding the services his firm could provide Honda and in part to the strong foundation of technical skills developed during its history as a machine shop and provider of advanced controls to other stampers.
Page26 Over eight years of experience with BP, Honda has gained a greater appreciation of the demands the BP process places on a supplier’s absorptive capacity. Accordingly, they now make sure the supplier is in a position to devote the necessary resources to the improvement process before beginning - no major product launches going on for Honda or another customen willing and able to devote at least one person full time, and with no major problems with its organizational structure, managerial resources, quality and production scheduling systems, or labor rekitions. According to Rick Mayo, who directs BP for Honti There are three things we look at now to evaluate whether BP will work @ their mfh@r@m; secon~ what else is on their plate; and thid who else is in there [other customers]. What we’d do now if a Capitol Plastics arose would be to do an analysis of their overaU management stmctum and resources. We would see that the problem is much bigger than BP. When you realize it’s more than BP, don’t e to solve it with BP. Before, we tried to make BP an answer for all issues. Now, we realize that a basic precondition of BP is having capable people in place. For example, suppose they fired a quality manager and don’t have a replacement. If we go @ we become the quality manager. I might send in a BP guy, but to manage quaMy, not to do BP. Mkr they’ve got the basic issues under control (we’re getting parts reliably, they change dies when they shoul~ they’re not promoting unqualified people), then we can do BP. Honda also felt that it was very important to encourage suppliers to see continuous improvement as part of their identity, by finding a way to make it fit with their organizational stmcmre, culture, and history and by striving for a consistency of approach across customers. According to Mayo: We’ve learned that we’ve got to get BP into their company so it’s not seen as a radical change. We used to meet with the top guy and say ‘do this project.’ Now we realize that the supplier needs to have their own way of doing BP. So we ask “what will fit best with your overall plans?” They don’t even need to call it BP. The improvement activities need to be part of their culture, their vision. BP is one club in the golf bag - it’s probably the driver, and we hit it hard - but it’s not meant to be everything to everyone. Thus over time, Honda baa refined its assessment of which suppliers should participate in BP and even more importantly, when and how to undertake a BP project at a supplier. Once suppliers have been selecte@ manage the relationship in a way that minimizes long-term dependenceand speeds the transition to self-sufficiency. One risk of a “hands-on” knowledge tmnsfm process is that the recipient firm (the supplier) may come to rely on the support services from the provider firm (the customer) and thereby limit its learning (AttewelL 1992). The cases offer some support for the idea that the suppliers might view BP as a service and become dependent on Hen@ thereby reducing their incentive to master lean production on their owm
Page27 Honda associates may have exacerbated the problem by their intensity about thing problems, which may have encouraged supplier employees to feel “unworthy” and to defer to the BP staff. (A manager at one supplier, not included among the cases in this paper, described in vivid detail the combination of relief and anxiety when the “snowflakes” -
white-uniformed Honda employees -
descended on the pro&ction
lines at his plant.) When Honda withdrew its BP staff, hoping that the supplier would institutionalize these processes, it was @uently
disappointed with the solo initiatives that followed.
This was particularly apparent at Capitol Plastics, where performance gains achieved on certain production lines diminished after the BP team lefl. subassembly
tasks were successfully
At Progressive, performance remained strong,
added, and technological
implemented But Honda remained somewhat hstrated
improvements
were effectively
with the limited depth of Progressive’s
management capabilities. According to one Honda manager, “We try to tell them they can’t stay as just a mom-and-pop shop and do big business. They can’t do it all with ‘promote-fkom-within’. At some point, they’ll need more professional management.”
In contrast, dependency was less of a concern at Tower.
At the beginning of their relationship, Tower was somewhat dependent on Honda’s technical assistance to work with unfamiliar Japanese dies and equipment. Yet this dependence was difficult for Tower, whose identity as a capable manufktumr was well-established. The ambivalence about being so dependent on Honda (and about the demands of the BP process) eventually servet we believe, to motivate Tower to become more self-sufllcient with respect to lean production capabilities and to synthesize what it was learning horn its two major customers. However, as Tower became more self-reliant, it may also have become less receptive to Honda’s “hands-on”
approach and more favorable towards Ford’s
“manage-by-the-numbers” approach. That the relationship between supplier and customer during the %arning”
period is often
characterized by tension and ambivalence is not necessarily a bad thing. It can motivate the supplier to push through the dependency stage and become self-retiant.’” Honda walked a fine line between making it easy for suppliexa @ to lean and creating an environment that encouraged leaning new capabilities. Honda’s investment of time and energy in helping suppliers was a visible sign of commitment something the suppliers could hold onto and me to jtiify contmctual commitments.
the “le~ of f~th” R-
in the ~s~ce
of
& noted above, BP also demonstrated the behaviors Honda hoped the
suppliers would adopt on their own.
Page28 b summary, reaching self-.su.i7iciencywith respect to newly-acquired lmowledge is more likely when the supplier has a moderate degree of identification with and dependency on the customer. If these are too II@ the supplier will be tempted to continue to rely on the customer for assistance. If they are too low, the supplier will feel that the customer has nothing to teach. On the basis of the limited evidence from these cases, we conclude that Honda has achieved the most supplier self-reliance with larger companies, who have an identity as strong, competent actors and thus try to reduce their dependence on Hon& as quickly as possible. Balance “learning” and “monitoring” behaviors, while being sure to generate supplier trust.
Customers who want their suppliers to improve must balance the need to monitor the suppliers’
existing pdormance that pdormance
while encouraging them to learn new skills which in the short term might disrupt
(Sabe~ 1993). Indee~ if supplier capabilities change sufficiently, it could force changes
in the monitoring system.
While BP was primarily a mechanism for learning, it did allow some
monitoring. However, tensions over monitoring at times seemed to threaten learning, as was the case at Tower. For the most part, Honda seemed willing to let BP to stand alone as a learning mechanism and to deal with monitoring as part of routine interactions between Honda purchasing, Iinancid
and quality
fhllctiOIISand each su@ier. While some tension between learning and monitoring is inevitable, the more capable suppliers become, the more they can participate in discussions in which both sides benefit, as occumed in the ihitfid combination of US and Japanese die practice at Tower and Progressive.
Honda appeared to
recognize the benefits of such discussions, becoming more flexible in its approach to BP over time, with more willingness to blend a variety of activities together, even those not part of the initial BP plan Honda also came to realk
that suppliers might undertake separate initiatives with other customers or outside
consultants, and that these too should be treated as potentially complementary to BP.L* Don’t worry about knowledge spillover to competitors through a shared supplier. QQworry about impact of multiple customers on a supplier’s responsiveness.
As
we
argue below,
lean
production is certaidy beneficial in seining other customers besides Honch Honda managers seemed to take a “rising tide lifts all boats” view of the risk that their rivals might free-ride on their investment in knowledge transfm. “If Ford and GM get better, that’s OK. That will push us to get better.” Also, unlike tiorrnation
about fbture products or specific technological advances (which Honda
certainly did protect), Honda managers seemed to believe that efforts to eliminate the spi.llover of
Page29 knowledge about lean production to competitors might run the risk of suppliers gaining only some of the benefits that accrue to systemic implementation. Indeed helping a supplier achieve consistency across all its operations and all its customers was an important component of Honda’s strategy for supplier self-reliance. Mayo told us: Measurement is one of the biggest things we do - measuring everything in the plant in the same way, gathering real data fkom physical operations. That how we end up working on Fo~ GM, and Chrysler lines as well as our own. We try to make continuous improvement part of the supplier’s culture. Furthermore, Honda’s emphasis on long-term, “marriage’’-like relationships with their key suppliers may also mitigate the risks of lmowledge spillover.
The loyalty generated by a SUCCCSS&I
knowledge transfm process can mean that suppliers remain more responsive to their teacher even over many decades.’2 We want to help suppliers solve their biggest problems, but also to think long-run and to be self-reliant. We train our BP staff to pay attention to issues of philosophy and commitment. That’s the strength of BP C it starts in one place and leads to other things. Honda seemed more concerned about the issue of supplier responsiveness to its needs in the face of competing customer demands.
According to one Honda manager, “Some of these suppliers get
pressure fivm many customers and ofkn say ‘yes’ to everything.
But they have limited managerial
resources for dealing with all those demands.” BP activities are increasingly postponed for suppliers with “too much on their plate,” as noted above, which most commonly occurs when a supplier is launching new products for more than one customer. This concern also seemed to provide the rationale for Honda’s extensive supplier development activities at the best smaller suppliers, despite the greater difficulty in helping them achieve self-reliance.
Lessons for Sappliers Understand what skills will be taaghg how long it will take to learn them, and how applicable they will be to other customers. Should suppliers be willing to get involved with a customer bent on improvement - theb?
Certain advantages of a knowledge transfm process like BP seem clear - the
operational improvements tim
the actual BP projects, the development of internal capabilities for
transferring these improvements to other plants and divisions, the strengthened important customer.
relationship with an
Yet a skeptic might argue that it is a mistake for a supplier to make major
Page30 investments in building capabilities that suit a customer’s requirements in the absence of solid long-term contmctual agreements about fixturebusiness. If a skill is specific to one customer, then a supplier should make sure in advance that customer is *to
~
the SUPP~CX of a r-
on its ~v-ent
in iefing-othtie
the suppiier is vulnerable
to a ‘hold-up’, in which the customer does not allow the firm a return on its investment, knowing that it has no alternative use for the asset (Klek
Crawfort and Alchi~
1978). If the skill is general, but
requires a long gestation period then the supplier should ensure that it can weather the learning period, either due to commitment from the customer or from retained earnings. BP seems to fall in this second category; it teaches general problem-solving skills helpfid to a variety of customers. Even the BP-related investments in capital equipment that we saw were for general~urpose machines. However, in the short term, the intense focus on one line required by Honda can be a &traction tkom meeting the needs of other customers. This tension was expressed most clearly by Tower, which was cautious about allowing its intensive relationship with Honda to hurt its position with Ford by diverting too much energy away born developing the capability and relationships necessary to obtain a major design role. Make sure your customer is trustworthy. Sako ( 1992) distinguishes among three types of trust: competence trust (a belief that the customer is capable of doing what it says it will do); contmctual trust (a f=ling that the customer will abide by its agreements), and goodwill trust (a belief that the customer will take initiatives for mutual benefit, and refkain horn unhir advantage-taking). AU types of trust are important for the success of an intensive knowledge-transfer project. Fi@ it is important that the customer be a competent teacher. Even if the customer doesn’t charge for insmctioq
the process of learning a general skill like lean production can still be quite costly
in terms of time for supplier managers, engineers, and employees. Regardless of their other reactions to BP, the suppliers were uniform in their respect for Honda’s mastery of lean production philosophies and methods - ofkn called the “Honda Way” - particularly as manifested in the skills of the BP coordinators sent to their plants. This sense of Honda’s high level of competence was crucial for overcoming the initial skepticism of many suppliers towards the BP process. Second contractd
trust - which on the face of itj wouldn’t apply to Honda because it rarely
provides formal contracts - remains extremely important. Knowing that a customer abides by unwritten contracts (e.g. Honda’s implicit guarsntee of continued annual business for responsive and capable
Page31
suppliers,at a level that at least matches the previous year’s volume) can strengthen a customer-supplier relationship more than seeing an customer in an arm’s-length relationship go “by the book” in obsemi.ng the terms of a formal contract. Thir~ because the knowledge transfer process opens up so many unforeseen avenues for improvement (and potential investment), it is crucial that the supplier believe that its customer is trustworthy in a goodwill sense. The intense amount of information-sharing and coordination between suppliers and a customer like Honda is often resisted at &t, because of fear that this information will be used opporhmistically .13 Yet suppliers often change their views if they see that the customer is merely seeking possible process improvements. Furthermore, as shown in the Tower and Progressive cases, firms can develop goodwill trust over time, progressively moving towards more and more mutual responsibility, interdependence, and open information exchange over time. A knowledge transfer program like BP can be m impotiant part of the trust-building process, since the customefs ~v~tment
in temh~g a supp~er E
specific to that suppl.ieq if the firm no longer supplies to the customer, the customer loses the benefit of having provided the training. Provide incentives for all members of the firm to contribute their ideas. Just as supplier management needs to trust customer management, supplier workers need to trust their managers, on both the competence and the goodwill dimensions. This was a difficult part of the BP process, at least in part b-e
supplier managers underestimated the importance of the changes required in labor-management
relations.
Capitol unleashed an initial flood of employee enthusiasm, which turned to a trickle when
management dragged its feet on the implementation of employee suggestions. In contrast, Progressive’s management responded more favorably towards suggestions, particularly from its more skilled employees, and reinforced the emphasis on problem-solving by providing extensive employee training, both on-the-job and at outside educational institutions. Overall, we found that Honda’s emphasis on the impoti
of shop-floor employee involvement beginning with the very first BP project and its
requirement that no workers be laid off due to BP-related efficiency gains helped shift managerial mindsets at the suppliers we studied. Grow your capabilities as fast as you grow your business — if not faster. The experience of becoming a supplier-partnerfor a tlist-growing firm like Honda was, for the companies we studie~ ~
md ~
= well as Stimuhing and exhilarating. Proving capable of handling Honda’s
demands was ofhm a ticket to gaining additional business from the other Japanese transplants and the Big
Page32 Three. In the fhce of such opportunities for volume, it could be easy for suppliers to defm investments in
fixturecapabilities
- whether in process technology, information systems, managerial expertise, or
production wo*er skills.
However, such a defexnl is, from Honda’s perspective, the most serious
impedimentto umtinuing an effective relationship with a supplier. According to Rick Mayo, We don’t want companies to grow their business faster than they grow their capabilities. We teIl suppliers to grow at the right pace to be ready when the time comes for the next phase of work together. For example, when we want to get a supplier involved in the “guest engineer” program (where the supplier sends engineers to work on “design for mada@r@’ issues with Honda’s product designers), they need to have the right kind of engineers already on board and t%rniliarwith their operation.
We view self-reliance in suppliers as a means to achieve long-term competitiveness. For most of our competitors, when they do supplier development activities, it’s just about today. We care about today but we also want residual benefit for the future. Some suppliers don’t get it but some get it very well. Those that get it are better suppliers for us and better suppliers for our other customers, which means we know they’ll be around. The larger firms are not necessarily better than the smaller firms at doing this, in Honda’s view, but they do pOSSeSscertain advantages - access to cspital, depth of managerial resources, a broader customer base and hence 16s vulnerability to fluctuations in Honda’s volume. The fti
that there is
widespread consolidation in the automotive parts industry, primarily through merger and acquisitio~ indicat= that many investors, customers, and supplier managers believe that size is a perquisite survival in the global sourcing competition of the tie.
to
But we would offer a cautionary note about the
dangers of suppliers growing too big, too rapidly.14 Size alone will not guarantee the successful development of the many capabilities required by lean customers like Honw
in fh@ growth through
diversification into new products and processes may deter the development of those capabilities. Honda catainly wants suppliers to develop their own capabilities because, as the BP experience shows, they are Iikcly to absorb Honda’sphilosophy and practical lessons most successfldly. But another benem from Honda’s pempective, is that these self-reliant suppliers ase speedier about %rning the right to disagree” and thus providing an impetus for Honda’s own improvement.
Page33 Conclusion Given the difficulties described above, should a customer still by to create lean suppiiers? Should suppliers agree to work with those customers pursuing this goal? We believe that both suppliers and customers can benefit fkom entering into the kinds of knowledge transfer arrangements described in this study of Honda’s BP. However, the specific mechanism chosen by Honda is far ffom the only way in which such knowledge transfer can take place.
We have observed different mechanisms at other
companies (we delineate several in a separate paper, forthcoming in Adler, Fruim and Liker, 1998) that are less %nds-on” situation”.
and time intensive, and less focused on learning from the “actual part, actual place, actual
While these mechanisms are arguably less effective at conveying the tacit knowledge
associated with lean productio% they might also be less costly to the customer and less disruptive in their challenge to the supplier’s identity. Some lmowledge transf= arrangements seem clearly undesirable in terms of imbalances in the risks and rewards that either customers or suppliers are asked to bear.
Where knowledge transfkr
mechanisms require highly customer-specific investments of time and capital without a long-term commitment ffom the customer, the risk for suppliers may be unacceptably high. On the other han~ if customers agree to detaild
legally-enforceable long-term contracts, they will lose much of their potential
leverage with respeet to getting suppliers to learn new capabilities. However, even the best transfm mechanism, applied to a highly absorptive and responsive recipient, is not sufficient to guarantee successful knowledge transfer. The fimdamental lesson of Honda’s BP experience is that a supplier~omer high trust bmeen
relationship which generates high motivation for learning and
provider and recipient is a crucial condition for any transfer of a complicate
tacit body of knowledge like lean production.
largely
Page34
a
References
Abernathy, Willianq Kim Clarlq and Alan Kantrow (1983). Books.
Industrial Renaissance. New York Basic
and Jefiiey Lilcer (1998). Remade in Ameriea Trausulanting and Transform.ing Adler, P~ MarkFJamnese Production Svstems. Forthcoming horn Oxford University Press. Attewe4 Paul (1992). “Teelmological Diflhsion and Organizational Learning,” Ormnhtion Vol. 3, No. l,pp. 1-19.
Science>
Bennet, James (1994). “Detroit Stmggles to Learn Another Lesson Iiom Japan”, New York Times, June 19, 1994. Coh~ Wesley M. aud Daniel A Levinthal (1990). “Absoqtive Capacity: A New Perspective on Learning aud hmovatio~” AdmhMrim “veSciences Ouarterlvj Vol. 35, No. 1, pp. 128-152. Cusumano, Michael (1984). The Jamnesc Automobile Indus~. Cambridge: Haxvard Univemity Press. Cusumano, Michael and Akira Takeishi (1991). “Supplier Relations and Supplier Management: A Survey of Japanese, Japanese-Transpla@ and U.S. Auto Plants”, Strategic Manamnent Jourr@ 12,563-88. Dos~ Giovanni and Bruce Kogut (1992). “National Specificities and the Context of Change: The Convolution of Organization and Technology,” in Dosi and Kogut (eds.) Country Competitiveness, Technolo~, and the (lx anizition of Work New York Oxford Universi@ Press. Dyer, Jeflkey and W. Ouehi (1993). “Japanese-style Business Partnership: Giving Companies a Competitive Edge,” Sloan Manaz ernent Review, 35,51-63. Evelan~ J.D. and Louis Toe (1980). “The Deployment of Technology,” in L. Tornat.zs~ and M. Fleischer, The Processes of Teehnolo~ieal Innovatio% Lexington MA: Lexington Boolca. Flori@ Richard and Davis Jenkins (1996) “Transfm and Adoption of Organizational Innovations: Japauese TranspIanta in the United States,” manuscript Carnegie-Mellon University. Gerladq Michael (1993). AUianee Capitalism. Berkeley, CA: University of California Press. Helper, Susan (1991). “Strategy and Irreversibility in Supplier Relations: The Case of the US Automobile Industry,” Bw”ness Histoty Review, 65, no. 4, Winter 1991, pp. 781-824. Helper, Susan ( 1992). “An Exit/Voice Approach to Supplier Relations,” Grabber, Gemo$ d. Embed&d Firm: On the Socio-Economics of Indusm”alNetworks. London Routledge, 1992.
The
Page 35
Kenney, Martin and Richard Florida (1993). Beyond Mass Production. New York Oxford University Press. Kle@ Benj@ R. CradoI@ and Armen Alchian (1978). “Vexticai Integratio~ Appropriable Quasi-Rents, and the Competitive Contracting Process”, Journal of Law and Economics. 21 (October), 297-326. Kogug Bruce and Udo Zander (1996). “What Firms Do? Omnization Science, Vol.7,No.5,pp.502-518.
Coordinating
Identity, and Learning,”
MacDuflie, John Paul (1997). “The Road to ‘Root Cause’: Shop-Floor Problem-Solving at Three Auto Assembly Plsnts”, Manwment Science, Voi. 43, No. 4, pp. 479-502. Mair, Andrew (1994). Honda’s Global Local Coloration: Martin’s -S.
Jammization the Honda Way. New York St.
Nelsoq Richard and Sidney Winter (1978). An Evolution arY Theorv of Economic Chawe, MA. Hamard Universi~ Press.
Cambridge,
Nishiguc~ Toshihiro (1994). Strate~ic Industrial Sourcimz: The Jauanese Advanta.ze. New York Oxford University Press. Od@ K., K.One, and F. Adachi (1988). The Automobile Industrv in Jar.)an:A Study of AnCillalYFirm Develo~ment. Oxford Oxford University Press. One, Taiichi (1988). Worldce
ManaEement. Cambridge, MA: Productivity Press.
Sabe~ Charles (1994). “’Learning by Monitoring’: The Institutions of Economic Development” in Neil Smelser and Richard Swedberg (eds.), Handbook of Economic Sociolo~. Princeto% N.J.: Princeton University press. &dco,Mari (1992).
priCCS.
Oudi
tv.
and Trust. Oxford: Oxford University Press.
Srni@ Michael ( 1991). Governance bv Trust. New York Columbia University Press. SzuhnalQ Gabriel (1995). “Appropriating Rents horn Existing Knowledge: I.ntra-FirrnTransfm of Best Practice,” PhD. Tk@ INSEAD, Fontaineble+ France. Womaclq James P., Daniel Jones, and Daniel Roos (1990). The Machine That Chawed the World. New York: Rawson Associates/MacMillan.
Page36
NOTES 1. For more&@ see a longer treatmentof our researchat Honda(forthcomingin Adler,Fru@ and Liker, 1998), which includesall six supplier casesand comparisonsof Honda’ssupplierdevelopmentactivitieswiththoseof other automotivecompanies.
2. We provided Hondawith the criteriafor supplierselectionand Hondaprepareda list of 15from whichwe chose the final set. Hondainformedthe suppliersthat they supportedour projectbut we madearrangementsfor the suppliervisits ourselvesand no Hondaemployeeswerepresentduringour visit.
3. Hondadoes steerits suppliersto goodpartners,in one sense,by askingtheir suppliersin Japanto assistthe U.S. supplierof the samepart. Whilemuchof this assistanceis technicalandproduct-specifiqsomeinvolvesmore generaladviceabout the productionsystem YettheseJapanesesuppliershavea limited interestin developing the capabilitiesof a U.S. supplierwho will be competingwith them forbusiness. Theyare likelyto provide someasistance for the sakeof goodwillin their relationshipwith Hen@ but not the long, intensivetutoring necessaryto inmrpomte !llllythe principlesof leanproduction.
4. It is mrikinghow ofte~ in the courseof our interviewsat Hen@ employees(frommanagemand engineersto purchasingstaff) would repeatphrasesattributedto Mr. Hondaas centralto the ‘Won&Way.” Examples(from Mair, 1994,p. 45) include: “Proceedalwayswith ambitionand youthfulness.Respeetsoundtheory. Develop freshideas and make the most effectiveuse of your time. Enjoyyour workand aiwaysbrightenyour working @WSP~. ~ve CO-Y for a ~OniOUS flOWof work. Be evmmiti of the valueof ~ch and endeavor.*’ 5. Ono(1988)foundthat experiencedemployeesoftenprovideinstant diagnosesof problemsin a situationthat seemsfamiliar,ernphasiziq “first-level”causes,e.g. “thatmachinealwaysbreaksdown whenwe run production at this volume”and solutions,e.g. “weneeda new machine.’*Going fmther requiresprobingfor whythe machinebreah down (e.g. “preventivemaintenancenot done”(why?);“becausemaintenancemenare busy repairinganothermachine”(why?);“becausethe othermachineoverheatsat high volumelevels”(why?); “beeauseit is locatedin a part of the plant with poor ventilationand no air conditioning”(why?);‘%ecauseit wasmoved thereto makeroom for a newpiece of equipmentthat neverarrived.” Thetrue “rootcause”of the obsemedproblempoints towardsa very differentsolution (movingthe machineor improvingventilationat its location)than the original diagnosis(replacethe machine).
6. At the time we visited this single-plantfirrq in 1992,it wasownedand managedby an entrepreneurwhohad purchasedit nine years earlier. Sincethat time, the fm has beenacquiredby a largeautomotivecomponents company,givenanew name,and put under new management. AUof the eventsin this casestudyprecedethe acquisitionand did not involvethe currentmanagementteam
7. MWtiofmfidd~k
19W,hf~w_%J. ToW. Itwassubsequently acquimdand mergedwith severalother medium-sizedstampinguxnpanies in the areato form“TowerAutomotive”. Evaything describedin this caseprecedesthe acquisition. For simplicity,we will refmto the company hereafk as ‘%wer”.
8. The casestudy of Donnelly, which we omit hereto meetlength requiremen~ is noteworthyas an exampleof Honda’s~ to help a capablesupplierlearnan entirelynew andunfamiliarproductionpin order to build on an @sting supplierrelationshipratherthan seekinga newsupplierwith that qertise. Donnelly initiauymadeqainted doormirrors for Hon& WhenHondadecidedto offerpainted mirrorson its new models it encoumgedDonnellyto equip its new plant for painting,a processDonnellyhad neverdone. The initial paint proca$ mostly manual,had major qualityproblemsso Hondi urgedDonnellyto investin a fully automate4 sophisticatedpaint line. Despitehaving no explicitcontractor commitmentsto purchasein hancL Donnelly%board did approvethe$5 million investment. Evenwith the newpaint line, intensiveBP activities with Hondawereneededdue to initial qualityproblems. Eventually,Donnellyovercamemost of theseeariy problemsandIwganto win painted minor businessfromother m compani~ boosting its corporatesalesfrom $20 million in 1990to $30 million in 1992.
Page 37
9. In 1996,Ruston Simon soid progressiveto anotherinvestor,althoughhe remainsas pre-sident. 10. Even onceachievc4 self-reliancecould havedifferentmeanings. For Hondaiself-reliancedid not mean substantiallyless of the intenseinteractionsbetweencustomerand supplier. For U.S.suppliers,in contrast,the idea of self-reliant%might be best expressedas Anowleaveme alone.” 11. In fact, Hondahas recentlytraineda team of engineersat the ClevelandAdvancedManufacturingProgxaq a technicalassistanceagenq, in BP techniquesso that they can help Hondameetthe demandfor the process among suppliersto Hon& and to other fm. 12. SeeOne, Adachi,and Odaka(1988)for examplesof long-termloyalty to Toyotain Japangeneratedin part by the automakers’technicalassistanceprograms. 13. %ukmski (1995)arguesthat trust betweenthe knowledgeproviderand the recipientis a key factorin motivatingthe recipientto learn whatis beingtaught;this motivationis as impomnt a factorin successful learningas is a moretechnicalfactorlike the amountof prior relatedknowledgea retipient has. 14. A similarsentimenthas been sound@ of late,by Big Threeexecutivessuch as Cluyslefs PurchasingDirector Tom StaUkamp:“Thereis somedangerthat suppliersare getting overlyintegrated- We weren’tvew good when we wereverticallyintegrated You should stickto whatyou do best.” (ClevelandPlain Dealer,August 18, 1996).
